Deep Learning

Invited Talk: Can Deep Learning Models be Trained with Annotations Collected via Crowdsourcing?

One of the major challenges facing researchers nowadays in applying deep learning (DL) models to Medical Image Analysis is the limited amount of annotated data. Collecting such ground-truth annotations requires domain knowledge (expertise), cost, and time, making it infeasible for large-scale databases. We presented a novel concept for training DL models from noisy annotations collected through crowdsourcing platforms, i.e., Amazon Mechanical Turk, Crowdflower, by introducing a robust aggregation layer to the convolutional neural networks. Our proposed method was validated on a publicly available database on Breast Cancer Histology Images showing interesting results of our robust aggregation method compared to baseline methods, i.e., Majority Voting. In follow-up work, we introduced a novel concept of an image to game-object translation in biomedical Imaging allowing medical images to be represented as star-shaped objects that can be easily embedded to readily available game canvas. The proposed method reduces the necessity of domain knowledge for annotations. Exciting and promising results were reported compared to the conventional crowdsourcing platforms.

Capsule networks against medical imaging data challenges

A key component to the success of deep learning is the availability of massive amounts of training data. Building and annotating large datasets for solving medical image classification problems is today a bottleneck for many applications. Recently, …

Deep autoencoding models for unsupervised anomaly segmentation in brain MR images

Reliably modeling normality and differentiating abnormal appearances from normal cases is a very appealing approach for detecting pathologies in medical images. A plethora of such unsupervised anomaly detection approaches has been made in the medical …

GANs for medical image analysis

Generating highly realistic images of skin lesions with GANs

As many other machine learning driven medical image analysis tasks, skin image analysis suffers from a chronic lack of labeled data and skewed class distributions, which poses problems for the training of robust and well-generalizing models. The …

Intraoperative stent segmentation in X-ray fluoroscopy for endovascular aortic repair

Multiple device segmentation for fluoroscopic imaging using multi-task learning

Scene coordinate and correspondence learning for image-based localization

Scene coordinate regression has become an essential part of current camera re-localization methods. Different versions, such as regression forests and deep learning methods, have been successfully applied to estimate the corresponding camera pose …

Weakly-supervised localization and classification of proximal femur fractures

When regression meets manifold learning for object recognition and pose estimation

In this work, we propose a method for object recognition and pose estimation from depth images using convolutional neural networks. Previous methods addressing this problem rely on manifold learning to learn low dimensional viewpoint descriptors and …